JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION …stone/PDFs/AWRA2001.pdf · JOURNAL OF THE...

JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATIONVOL. 37, NO. 5 AMERICAN WATER RESOURCES ASSOCIATION OCTOBER 2001

IMPACTS OF CLIMATE CHANGE ON MISSOURIRWER BASIN WATER YIELD'

Mark C. Stone, Rollin H. Hotchkiss, Carter M. Hubbard, Thomas A. Fontaine,Linda 0. Mearns, and Jeff G. Arnold2

ABSTRACT: Water from the Missouri River Basin is used for multi-ple purposes. The climatic change of doubling the atmospheric car-bon dioxide may produce dramatic water yield changes across thebasin. Estimated changes in basin water yield from doubled CO2climate were simulated using a Regional Climate Model (RegCM)and a physically based rainfall-runoff model. RegCM output from afive-year, equilibrium climate simulation at twice present CO2 lev-els was compared to a similar present-day climate run to extractmonthly changes in meteorologic variables needed by the hydrolog-ic model. These changes, simulated on a 50-km grid, were matchedat a commensurate scale to the 310 subbasins in the rainfall-runoffmodel climate change impact analysis. The Soil and Water Assess-ment Tool (SWAT) rainfall-runoff model was used in this study.The climate changes were applied to the 1965 to 1989 historic peri-od. Overall water yield at the mouth of the Basin decreased by 10to 20 percent during spring and summer months, but increasedduring fall and winter. Yields generally decreased in the southernportions of the basin but increased in the northern reaches. North-ern subbasin yields increased up to 80 percent: equivalent to 1.3 cmof runoff on an annual basis.(KEY TERMS: global climate change; surface water hydrology; Mis-souri River Basin; meteorology/climatology; regional circulationmodel; water resources; reservoir modeling.)

INTRODUCTION AND OBJECTIVES

Projected increases in concentrations of atmospher-ic carbon dioxide (C02) and other greenhouse gaseswill likely result in a changed climate. Climatechange will affect water availability in the MissouriRiver Basin. Water uses in this region include irriga-tion, flood control, navigation, hydropower, recreation,natural resources, and consumptive use: systems thatare often in direct competition for the same resources.

For example, crops and agricultural production coverapproximately 46 percent of the Missouri River Basin,of which 5 percent is irrigated (Srinivasan et al.,1994). These and other water resource needs makethe Missouri River Basin extremely vulnerable to anyhydrologic impacts of climate change.

A complete method has been developed for analyz-ing the impacts of climate change on water resourcesusing a continuous daily time step model that numer-ically routes the water yield through the MissouriRiver Basin. The Soil and Water Assessment Tool(SWAT) has been modified to incorporate data from aRegional Climate Model (RegCM).

The objectives of this paper are to describe a repro-ducible method for evaluating climate change impactson the Missouri River Basin and to analyze climatechange impacts on basin water yields.

PREVIOUS MODELING EFFORTS

There have been several simulation studies of thepotential effects of climate change on agricultural pro-ductivity for the Missouri River Basin, but only a fewincluded water resources. Rosenberg completed astudy on the Missouri, Iowa, Nebraska, and Kansas(MINK) region of the central U.S. Weather records ofthe 1930s 'dust-bowl' era were used as a surrogate cli-mate change scenario (Rosenberg, 1993). It was con-cluded that if the 'dust-bowl' climate was to recurtoday, flows from the Missouri and Upper Mississippi

'Paper No. 99180 of the Journal of the American Water Resources Association. Discussions are open until June 1, 2002.2Respectively, Graduate Research Asst. and Associate Professor and Director, Albrook Hydraulics Laboratory, Dept. of Civ. and Env. Eng.,

Washington State University, P.O. Box 642910, Pullman, Washington 99164-2910; Civil Engineer, Dodsons and Associates, 5629 FM 1960West, Suite 314, Houston, Texas 77069-4216; Assistant Professor, Dept. of Civ. and Env. Eng., SD School of Mines and Tech., 501 East SaintJoseph St., Rapid City, South Dakota 55701; National Center for Atmospheric Research, P.O. Box 3000, Boulder, Colorado 80307-3000; Agri-cultural Engineer, USDA-Agricultural Research Service, 808 East Blackland Rd., Temple, Tx 76502 (E-MaiLfHotchkiss: [email protected])

JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION 1119 JAWRA

Stone, Hotchkiss, Hubbard, Fontaine, Mearns, and Arnold

would be reduced by 28 percent. Hurd et al. (1996)used variants of a two-layer Variable InfiltrationCapacity (VIC-2L) hydrologic model to estimate theimpacts of a set of predetermined changes in meantemperature and precipitation on water yield in fourU.S. water resources regions including the MissouriRiver Basin. Three increments of temperature change(1.5, 2.5, and 5.0°C) and four increments of precipita-tion change (+15, 7, 0, and -10 percent) were applied.Changes in water yield ranged from +20.5 percent at1.5°C to -56.8 percent at 5.0°C (Rosenberg et al.,1999). Lettenmaier et al. (1999) combined VIC-2Lwith the output from three transient General Circula-tion Models (GCMs) and one double carbon dioxide(2xCO2) GCM to evaluate potential effects of climatechange on water resources in six river basins includ-ing the Missouri. Water yield decreases of 6, 24, and34 percent were reported for transient climate scenar-ios and a 2 percent increase was reported for the2xCO2 scenario.

Modeling Procedure

The assessment process may be described in threemodules: (1) a changed climate scenario was producedwith a GCM at a resolution of about 5 degrees inphysical space (Giorgi et al., 1998); (2) a Regional Cli-mate Model (RegCM) was used to downscale the cli-mate data to a horizontal grid point spacing of 50 km;and (3) a Geographical Information System (GIS) wasused to incorporate the RegCM data into a hydrologicmodel that was then used to assess impacts on wateryield for a selected historic period.

Hydrologic Modeling

Many models have been developed to assess theimpacts of climate change on water resources in vari-ous parts of the world. Simulation algorithms haveprogressed from empirical and statistical estimatorsto more physically based models (Jorgensen, 1996).Rainfall-runoff models like the Soil and Water Assess-ment Tool (SWAT) were developed for large-scalehydrologic modeling (Srinivasan et al., 1998; Arnoldet al., 1999). However, the SWAT model has not yetbeen tied to"nested" RegCM simulation results forsub-continental climate change impact analysis as itwas in this paper.

SWAT

SWAT is a continuous watershed scale model thatsimulates the major components of the hydrologiccycle on a daily time step. The hydrologic model isbased on a water balance equation for soil water con-tent as follows (Arnold et al., 1998):

SW = SWi + Pt - Qt - ET - - QR (1)

where SW is the soil water content for the currentday, SW1 is the soil water content for the previousday, P is precipitation, Q is surface runoff, ET is evap-otranspiration, SP is percolation or seepage, and QRis return flow.

Surface runoff is modeled using a modified SCSCurve Number approach (USDA-SCS, 1972). TheRunoff Curve Number (RCN) is a retention parameterthat varies according to soil type, la:nd use, cover, andwater content. The RCN in SWAT is updated on adaily basis according to soil water content. Previousresearch has shown that the increased model com-plexity of methods such as the Green-Ampt equationover the relatively simple Curve Number approachdoes not necessarily translate to improved accuracy(Loague and Freeze, 1985; , Beven, 1989; Wilcox etal., 1990).

Modeling algorithms for the major hydrologic pro-cesses in SWAT are shown in Table 1. SWAT also sim-ulates the agriculture-based processes of biomassproduction, plant growth, and the fertilization andtranspiration suppression effects of CO2 enrichment(Arnold et al., 1995). Modifications to the ET and radi-ation use efficiency algorithms account for the effectsof changing atmospheric CO2 (Arnold et al., 1998).

SWAT was developed by the Agricultural ResearchService (ARS) and the Texas Experimental Station(TES) for the Hydrologic Unit Modeling of the UnitedStates (HUMUS) project in response to the amendedResources Conservation Act (Srinivasan et al., 1994).SWAT has been linked to GRASS, a Geographic Infor-mation System (GIS), by an interface that sets water-shed boundaries and extracts necessary inputvariables from digital databases (Arnold et al., 1995).

Missouri River Dataset

SWAT is a watershed-structured program thatdivides the area modeled into topographically definedwatersheds. The Missouri River Basin model wasdivided using USGS-defined eight-digit watersheds(U.S. Geological Survey, 1987). The eight-digit water-sheds are the smallest cataloging unit used by theUSGS. The 310 eight-digit watersheds comprising the

JAWRA 1120 JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION

METHODS

Impacts of Climate Change on Missouri River Basin Water Yield

TABLE 1. Modeling of Hydrologic Processes in SWAT (Arnold et at., 1995).

Algorithm

Observations or First-Order Markov Chain Model

SCS Curve Number Method

Variable Storage Coefficient

Stage-Storage or Reservoir Operating Procedures

Storage Routing Combined With Crack-Flow Model

Mean Air Temperature and Soil Layer Thmperature

Kinematic Storage Model

Baseflow Period, Ground Water Storage, and Re-evaporation

Lane's Method

Penman-Monteith, Hargreaves, or Priestley-Taylor MethodsModified Universal Soil Loss Equation (MUSLE)

Missouri River Basin range in size from 2000 to 5000km2 and are shown in Figure 1.

Figure 1. The Missouri River Basin SeparatedInto the 310 USGS Eight-Digit Subbasins.

Each eight-digit watershed was partitioned into asmany as 15 additional subbasins based upon land useand soil composition resulting in 3,494 subbasins formodeling purposes. Input variables were extractedfrom the basin databases shown in Table 2 withGRASS GIS and an interface program linked toSWAT. Subbasin topography was defined using USGSDigital Elevation Model (DEM) databases. Channeldimensions, slope-lengths, and other topographicparameters were estimated from these data.

Weather Components

SWAT uses six daily time step weather variables:precipitation, maximum and minimum temperature,wind speed, solar radiation, and relative humidity.Precipitation and temperature can either be historicor stochastic. Historical precipitation and tempera-ture data for the 1965 to 1989 simulation period wereread from existing data files. Data for wind speed,solar radiation, and relative humidity are availablebut are generally incomplete over the simulation peri-od. Therefore, statistical parameters were calculatedand these variables were stochastically generated(Arnold et al., 1995). All stochastically generatedinput was saved, resulting in one basic input datasetfor the study (Hubbard, 1998).

Representative precipitation was found for eachwatershed by the Thiessen polygon method(Wanielista, 1990). Weather inputs are specific to eachUSGS eight-digit in the Missouri River Basin. Thus,there are 310 sets of weather data files correspondingto the 310 USGS eight-digit watersheds that form theMissouri River Basin. Each subwatershed within theeight-digit watersheds shares the same weather datafile.

Incorporating the Main Stem Reservoirs

The U.S. Army Corps of Engineers (USACE) oper-ates a series of six main stem reservoirs in the upperMissouri River Basin (Figure 2). The reservoirsare extremely important because they control the55 percent of the area upstream from the reservoirsand are capable of storing up to three times themean annual runoff from this area. The reservoirs areoperated to meet the prioritized purposes of flood


Process

PrecipitationSurface Runoff

Channel Routing

Reservoir RoutingPercolation

Snowmelt

Lateral Subsurface Flow

Ground Water Flow

Transmission Losses

EvapotranspirationSediment Yield


TABLE 2. Digital Databases Used as Input in Assembling Models.

Database or Data Source Used For

USGS Digital Elevation Models (DEMs) Topography, Basin Delineation

USGS Land Use/Land Cover Land Cover

USGS Maps Hydrography

STATSGO Soils Data

Census of Agriculture, Department of Commerce City Map Crop Type

Tillage Residue Database Cropping Practice

Shallow Aquifer Baseflow Period Map Aquifer Data

control, irrigation, downstream water supply andwater quality, navigation and power, and fish andwildlife as set forth in the Master Manual (U.S. ArmyCOE, 1979).

Figure 2. The Six Missouri River Main Stem Reservoirs.

The original SWAT code algorithms determinedreservoir releases using simple uncontrolled spill-ways, obviously insufficient for modeling releasesfrom the main stem Missouri River reservoirs. Jor-gensen (1996) implemented new algorithms to simu-late historical releases based upon seasonalconsiderations and system priorities. A complete dis-cussion of the algorithm development and USACErules of operation can be found in Jorgensen (1996)and Hotchkiss et al. (2000).

Climate Modeling

The climate change scenario for this project wasdeveloped using a nested regional modeling techniquewhereby the RegCM was nested within control anddoubled CO2 runs of the Commonwealth Scientificand Industrial Research Organization (CSIRO) GCM(Watterson et al., 1995). The GCM was run with aspatial resolution equivalent to 3.2 degrees latitude(400 km) by 5.6 degrees longitude (500 1cm). The ini-tial and lateral boundary conditions for the RegCMwere provided from the CSIRO GCM at eight-hourlyintervals; these data were linearly interpolated to theregional climate model time step of three minutes.The RegCM was run at a spatial resolution of 50 km.A description of the nested modeling runs used in thisproject is found in Giorgi et al. (1998).

Model selection is important, as variations in cli-mate change predictions will impact study results.The CSIRO GCM was selected as the nesting modelover the study region because of its relatively highdegree of accuracy when used to estimate current cli-mate. The CSIRO model produced a very high qualitysimulation over North America (Giorgi et al., 1998).Predicted changes in climate by the CSIRO GCM areconsistent with models reviewed by the Intergovern-mental Panel on Climate Change Second AssessmentReport (IPCC, 1995). Great uncertainty exists in theaccuracy of regional GCM results. However, CSIROGCM results are similar to many regional results forNorth America regarding percentage change in pre-cipitation and range of temperature change.

This type of scenario was chosen for the study inlarge part due to its high spatial resolution. Given theresolution of the sub-basins making up the studyarea, using a climate change scenario on a spatialresolution similar to that of the sub-basins is muchpreferred, and may result in more physically mean-ingful results. Scenarios developed from these runs ofthe RegCM have also been used in several agricultur-al assessments (Brown et al., 1999; Mearns et al.,



1999; Mearns et al., 2001). Mearns et al. (1999)recently established that the resolution of climatechange scenarios can effect the determination of theimpacts of climate change on simulated regional cropyields in the Central Plains of the U.S. Mearns et al.(2001) also found that the scale of climate changescenario substantially affected the simulation ofchanges in crop yields on various levels of spatialaggregation in the southeastern U.S. Stone (2000)found that changes in predicted water yields as aresult of climate change were dependent on climatechange model resolution.

Implementing Climate Change Data

The RegCM was run at a 50 km grid point spacing;thus the Missouri River Basin is covered by a matrixof 28 by 42 grid points. The northern border of theMissouri River Basin coincides with the north bound-ary of the usable grid points (outside the interpolationregion on the borders). The impact of boundary effectswere found to be negligible in this study (Hubbard,1998). The RegCM produced output for each gridpoint in the entire region. Regional model grid pointswere correlated with the USGS eight-digit watershedsfor climate change modeling purposes (see Figure 3).

Figure 3. Missouri River Basin GCMand RegCM Grid Spacing.

The RegCM simulates the six variables needed todrive the hydrologic model. The model was run forcontrol conditions (with 330 ppm C02) and for dou-bled CO2 conditions (660 ppm). Five years of each runwere completed. From daily regional model output,five-year monthly averages of the variables used bySWAT were produced.

The RegCM for the changed climate conditions sim-ulated increases in both temperature and precipita-tion. Seasonal warming of 4 to &C and precipitationincreases of 6 to 24 percent were produced. Thelargest changes were found in the northern part ofthe river basin. Detailed descriptions of the regionalmodeling results can be found in Giorgi et al. (1998)and Mearns et al. (1999).

Creating the Climate ChangeScenarios for Input into SWAT

The 2xCO2 scenario for input to SWAT was createdby combining the differences or ratios from themonthly 2xCO2 and control run outputs with thebaseline climate data set. Differences were used fortemperature and ratios for all other variables. Exam-ples of RegCM results as applied to three selectedbasins are shown in Figure 4. Precipitation ratios aresignificantly greater in the Upper Missouri Basin,and are accompanied by higher temperatures eachmonth of the year. Temperatures are also warmer asapplied to the Upper Niobrara basin and the LowerMissouri, but precipitation ratios are much closer toone.

Water Yield Changes

Water yield changes are summarized by three-month seasonal averages with winter beginning inJanuary. Average water yields for each season fromeach eight-digit watershed for the 1965 to 1989 periodwere calculated from the extracted data for the base-line run and for the 2xCO2 run. The percent changesin water yield from baseline conditions were calculat-ed using the GRASS GIS interface and are shown inFigures 5 through 8.

Model water yields for the spring, summer, and fallseasons show more distinct spatial trends than wateryields for the winter months. SWAT simulationresults exhibit dramatic increases in water yieldsacross the northern and northwestern portions of theMissouri River Basin in the spring, summer and fall.Water yields in these areas increase by 70 percentand much more in certain eight-digit subbasins. Thesoutheastern portions of the basin display an overalldecrease in water yields during these three seasons.Most of the southeastern USGS eight-digit water-sheds show decreases in water yields of less than 20

MODELING RESULTS

meterS.200 0



Figure 4. Precipitation Ratios and Changes in Maximum Thmperature for Three Subbasins.

percent but some decrease by 80 percent or more.In spring and summer the decreasing water yieldsfrom the southeastern portions of the Missouri RiverBasin are enough to lower the total water yield of thebasin by 10 to 20 percent. Water balance summariesfor principal tributaries are shown in Figure 9 andportray the large increase in runoff in the northernbasins and decreased runoff in the south.

Main Stem Reservoir Changes

The total main stem reservoir storage changes dra-matically from baseline conditions (Figure 10). Thedifference between the releases from the reservoirsystem for baseline and 2xCO2 simulations can beused for mitigating the negative impacts of climatechange or optimizing the benefits. The difference inthe releases from the reservoir system represent thedifference in available water to the system. If releasesfrom the reservoir system are higher for the 2xCO2climate situation than for the baseline climate situa-tion, the additional releases represent extra waterthat is available for use for the 2xCO2 climate condi-tions.

The average reservoir system releases for the base-line and 2xCO2 simulations are 670 m3/s and 1160

m3/s respectively, representing 3.5 and 4.8 cm ofrunoff over the entire Basin. Planned increases inirrigation projects for the northern half of the basinhave yet to be implemented. If the projects were built,there would be an additional 2.4 cm of water availablefor the irrigated area. Basinwide, because less than5 percent of the entire basin is under irrigation, theincrease in water available for all irrigated lands inthe entire Missouri River Basin is approximately 24cm annually.

DISCUSSION AND CONCLUSIONS

The method of analyzing the impacts of climatechange on the spatial and temporal distribution ofwater resources outlined in this project is an excellenttool for exploring the potential impacts of climatechange scenarios and possible mitigation schemes.Water resources availability and distribution andagricultural production are vital to the plains regionof the United States. Therefore, a method of analysis,such as this, that evaluates impacts of climate changeon water resources across the region, is an extremelybeneficial tool for developing management strategiesto deal with climate changes. Moreover, care has been


iii.rliii l•••1 liiilIII !1•1..1111J.11j 11111

________ .• _-, .—II 111111111111111111-w-J-—


Figure 6. Percent of Change in Water Yield From Baseline Conditions for Spring (April through June).



Figure 8. Percent of Change in Water Yield From Baseline Conditions for Fall (October through December).



taken to match the scale of climatic input from theRegCM to the scale of the SWAT model.

Despite continuous advancements in the areas ofclimate and hydrologic modeling, several factors stilllimit the predictive capability of these models. Thisprohibits the use of model results as deterministicoutcomes and focuses attention on the idea of strate-gizing and scenario modeling. Hydrologic models areincreasingly more complex and capable of modelinglarge regions in detail. This capability is in part limit-ed by the predictive capabilities of the climate modelsthat provide meteorological inputs to the hydrologicmodel. GCM and RegCM models are considered theideal approach to climate modeling because predic-tions are based on the physical laws governing atmo-spheric processes. As computing capabilities grow andassumptions that limit the accuracy of climate modelsbecome unnecessary, GCM and RegCM models willbecome increasingly useful as predictive tools for cli-mate change.

Results of this analysis show an overall decrease inthe surface water yields from the Missouri RiverBasin. The spatial pattern of changes in predicted

Figure 10. Volume Change and Percent Change in Average Daily Cumulative Reservoir Storage.


80

60

40

20

0

-20

-40

85

58

Yellowstone Cheyenne tafle

River

Figure 9. Percent of Change in Average AnnualFlow for Six Tributaries.

5.0

4.5

— 4.0I-ww53.5C)

C.)C)I0

2.5SC)

2.0C)

1.5> 1.0

0.5

0.0

100

90

80

70

C)ou 0)

C)50CC)C.)1C)'IV

30

20

10

0

0 25 50 75 100 125 150 175 200 225 250 275 300

Month (1 965-1 989)


water resources is heavily influenced and related tothe spatial patterns of the predicted changes in cli-mate conditions. Climate variables in the northernregion of the Missouri River Basin change the most inthe predicted climate change regime. Precipitationand temperature both increase more in the norththan anywhere else in the Great Plains (Giorgi et al.,1998). Thus, the north also experiences the greatestchanges in predicted water yields and agriculturalproduction.

Model results indicate the reservoirs releaseenough additional water under double CO2 climateconditions to provide 24 cm of additional water for allirrigated areas in the Missouri River Basin annually.The average annual release from the reservoirs forthe 2xCO2 simulation is nearly double the averagerelease of the baseline simulation. An additional 1.57million hectare-meters of water are available from thereservoirs for each year of the 2xCO2 simulation. Theextra water can be diverted from the reservoirs forirrigation of the upper Missouri River Basin or reser-voir releases can be increased and the water can bewithdrawn downstream from the reservoirs to supple-ment irrigation of the lower half of the Missouri RiverBasin.

Results from this study are similar to thoseobtained in previous modeling efforts. The 10 to 20percent reductions in basin water yields are similar tothe study results by Rosenberg et al. (1999), Hurd etal. (1996), and the transient results of Lettenmaier etal. (1999). It is interesting to note that the 2xCO2 sce-nario completed by Lettenmaier produced a 2 percentincrease in water yield. This is likely due to varia-tions in climatic inputs and hydrologic modeling tech-niques.

ACKNOWLEDGMENTS

This research was funded entirely by the National Institute forGlobal Environmental Change through the U.S. Department ofEnergy (Cooperative Agreement No. DE-FCO3-90ER61010). Anyopinions, findings, and conclusions or recommendations expressedin this publication are those of the authors and do not necessarilyreflect the views of the DOE. Support was also received from theCenter for Infrastructure Research at the University of Nebraska-Lincoln.

LITERATURE CITED

Arnold, J. G., R. Srinivasan, R. S. Muttiah, and P. M Allen, 1999.Continental Scale Simulation of the Hydrologic Balance. Jour-nal of the American Water Resources Association 35:1037-1051.

Arnold, J. G., R. Srinivasan, R. S. Muttiah, and J. R. Williams,1998. Large Area Hydrologic Modeling and Assessment — Part 1:Model Development. Journal of the American Water ResourcesAssociation 34:73-89.

Arnold, J. G., J. R. Williams, R. Srinivasan, and K. W. King, 1995.Soil and Water Assessment Thol (SWAT) Users Manual. USDA-ARS, Temple, Texas.

Beven, K., 1989. Changing Ideas in Hydrology— The Case of Physi-cally-Based Models. Journal of Hydrology 105(1):157-172

Brown, R. A., N. J. Rosenberg, W. E. Easterling, C. Hays, and L. 0.Mearns, 1999. Potential Production and Environmental Effectsof Switchgrass and Traditional Crops Under Current andGreenhouse-Altered Climate in the MINK Region of the CentralUnited States. Ecology and Agricultural Environment.

Giorgi, F., L. 0. Mearns, C. Shields, and L. McDaniel, 1998. Region-al Nested Model Simulations of Present Day and 2xCO2 Cli-mate Over the Central Plains of the United States. ClimaticChange 40:457-493

Hotchkiss, R. H., S. F. Jorgensen, M. C. Stone, and T. A. Fontaine,2000. Regulated River Modeling for Climate Change ImpactAssessment: The Missouri River. Journal of the American WaterResources Association 36(2):375-386.

Hubbard, Carter M., 1998. Hydrologic Modeling of the MissouriRiver Basin in a Climate Change Model. MS. Thesis, Universityof Nebraska-Lincoln, Lincoln, Nebraska.

Hurd, B., M. Callaway, and J. Smith, 1996. Economic Effects ofClimate Change on U.S. Water Resources. Report prepared forElectric Power Research Institute, Palo Alto, California.

IPCC (Intergovernmental Panel on Climate Change), 1995. Inter-governmental Panel on Climate Change Second AssessmentReport. The World Meteorological Association and The UnitedNations Environmental Program.

Jorgensen, S. F., 1996. Hydrologic Modeling of Missouri RiverReservoirs in a Climate Change Model. Master's Thesis, Univer-sity of Nebraska-Lincoln, Lincoln, Nebraska.

Lettenmaier, D.P., A. Wood, R. Palmer, E. Wood, and E. Stakhiv,1999. Water Resources Implications of Global Warming: A U.S.Regional Perspective. Climatic Change 43(3):537-579, 1999.

Loague, K. M. and R. A. Freeze, 1985. A Comparison of Rainfall-Runoff Modeling Techniques on Small Upland Catchments.Water Resources Research 21(2):229-248.

Mearns, L. 0., W. Easterling, C. Hays, and D. Marx, 2001. Compar-ison of Agricultural Impacts of Climate Change Calculated fromHigh and Low Resolution Climate Model Scenarios: Part I. TheUncertainty of Spatial Scale. Climatic Change (in press).

Mearns, L. 0., T. Mavromatis, E. Tsvetsinskaya, C. Hays, andW. Easterling, 1999. Comparative Responses of EPIC andCERES Crop Models to High and Low Resolution ClimateChange Scenarios. J. Geophys. Research [Special issue on NewDevelopments and Applications with the NCAR Regional Cli-mate Model (RegCM)i 104(D6):6623-6646.

Rosenburg, N. J., 1993. The Mink Methodology: Background andBaseline. Climatic Change 24:7-22.

Rosenburg, N. J., D. J. Epstien, D. Wang, L. Vail, R. Srmivasan,and J. G. Arnold, 1999. Possible Impacts of Global Warming onthe Hydrology of the Ogallala Aquifer Region. Climatic Change42:677-692.

Srinivasan, R., J. G. Arnold, R. S. Muttiah, C. Walker, and P. T.Dyke, 1994. Hydrologic Unit Model for the United States(HUMUS). In: Advances in Hydro-Science and Engineering, Vol.I, Part A.

Srinivasan, R., T. S. Ramanarayanan, J. G. Arnold, and S. T. Bed-narz, 1998. Large Area Hydrologic Modeling and Assessment,Part II: Model Application. Journal of the American Associationof Water Resources 34(1):91-101.

Stone, M. C., 2000. Water Yield Responses to High and Low SpatialResolution Climate Change Scenarios in the Missouri RiverBasin. M.S. Thesis, Washington State University, Pullman,Washington.



U.S. Army Corps of Engineers (COE), 1979. Missouri River MainStem Reservoir System Reservoir Regulation Manual, MasterManual. Omaha, Nebraska.

U.S. Department of Agriculture, Soil Conservation Service (USDA-SCS), 1972. National Engineering Handbook, Hydrology Section4, Chapters 4-10. U.S. Government Printing Office, Washington,D.C.

U.S. Geological Survey, 1987. Hydrologic Unit Maps. Water SupplyPaper 2294. U.S. Government Printing Office, Washington,D.C.

Wanielista, M. P., 1990. Hydrology and Water Quantity Control.John Wiley and Sons, New York, New York.

Watterson, I. G., M. R. Dix, H. B. Gordon, and J. L. McGregor,1995. The CSIRO Nine-Level Atmospheric General CirculationModel and Its Equilibrium Present and Doubled CO2 Climate.Aust. Met. Mag. 44: 111-125.

Wilcox, B. P., W. J. Rawls, D. L. Brakensiek, and J. R. Wight, 1990.Predicting Runoff From Rangeland Catchments: A Comparisonof Two Models. Water Resources Research 26(10):2401.


JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION …stone/PDFs/AWRA2001.pdf · JOURNAL OF THE...

Documents

Transcript of JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION …stone/PDFs/AWRA2001.pdf · JOURNAL OF THE...